Efficient Frontier
85TrustedInstructions onlyOrchestrate complex tasks by delegating mechanical work to faster, cheaper subagents.
What it does
Efficient Frontier changes how your assistant handles large-scale operations. Instead of performing every step with a high-cost frontier model, this skill organizes the work into segments. It positions the main model as a central orchestrator and quality reviewer while assigning time-consuming, repetitive, or data-intensive sub-tasks to smaller, more economical agents.
This approach ensures expensive processing power is saved for high-level decision-making, such as architectural planning, synthesizing research, and final verification. By isolating bounded tasks, the assistant can manage broader project scopes while maintaining a consistent standard of accuracy and output quality.
When it earns its place
Reach for it when…
Use this when you face complex coding projects, large-scale repository research, or extensive debugging where the work can be split into parallel threads. It is ideal for scenarios involving numerous files or deep searches where the assistant needs to analyze multiple theories simultaneously before consolidating the final strategy.
Look elsewhere when…
Do not use this for simple, linear tasks where the overhead of spawning subagents exceeds the time required for a single, direct response. It is not suitable for creative writing or brainstorming sessions that rely on a single, continuous flow of logic and tone rather than discrete, verifiable segments.
What to say to use it
You never invoke a skill directly — the assistant picks it up when your request matches what it covers. These are the kinds of things that do that.
- help me manage this large refactor using subagents
- split up this repository investigation into separate parts
- orchestrate a parallel search for these bugs
- coordinate several agents to test these features
Install it
Install for all your projects
Copy the skill directory into your personal skills folder. Claude picks it up on its own the next time your request matches what it does — there is nothing to enable.
git clone --depth 1 --branch main https://github.com/BuilderIO/skills.git /tmp/skills
mkdir -p ~/.claude/skills
cp -r /tmp/skills/skills/efficient-frontier ~/.claude/skills/efficient-frontierInstall for one project only
Put it in the project's own skills folder instead, so it can be committed and shared with everyone working on that repository.
cp -r /tmp/skills/skills/efficient-frontier .claude/skills/efficient-frontierIf a skill of the same name exists in both places, the personal one wins.
Installing on Claude.ai or through the API instead? The full guide covers every surface.